various glue's strength in mortise joints

Re: various glue's strength in mortise joints

Ed Snow

>We did a test here at C.R. and foound that in a near perfect mortice and tenon joing or lap joint that normal white glue elmers or white craft glue is esentually as strong at things such as epoxy and polyurathane glue. Hide glue was not tested. We maesured in foot pounds torsion.

Re: various glue's strength in mortise joints

Adam Cherubini, NJ

>Bill,

This is interesting data, but here in the hand tool world, I think the gaps should be more like .03", not 8 mils. I'd be happy to get joints with only 8 mils of slop! More often than not I see .03" - .06".

My guess is that you would see the PVA strength fall off rapidly as gap increases. Conversely, hide glued joint strength should theorhetically increase as gap increases up to some point.

I originally thought this would make a great FWW article.

I was interested in learning:

1) What's the best glue for a gappy joint?

2) Since you really can't clamp a dovetail, are you better off making them gappy and using hide glue, or making them tight and using PVA? I make mine tight and use hide glue and I think that's the wrong answer.

3) If hide glue is gap filling, what's the upper bound? When do you have to shim that mortise and tenon?

But after my last encounter with FWW editors, I doubt they would be interested in such things; especially if an advertiser like Franklin got a drubbing.

Adam

Hand cut versus machine cut

Brent Langdon, Sterling VA

>I have wondered about the relative strength of hand cut mortises versus machine cut (router or mortiser). Compared to a router cut mortise, a hand chopped one is very rough. Like Adam points out, maybe hide glue would yield a stronger joint for a hand cut mortise and PVA would be a better choice for machine cut.

- Brent

Re: Hand cut versus machine cut

joel

>Depends on how you cut them.

If you pare the sides to final dimension as I do you get dead smooth sides trivially and a very tight fit. At least as good as with a router - maybe better as it's easier to gauge the final dimension to exactly what the tenon is and I'm not restricted to router bit sizes.

note:

if there is any doubt on a mortise strength - drawbore pin the mortise. This also solves any loosening up due to weather changes. In general anything that has a lot of stress on it such as doors should always be pinned. Pins can also be concealed too. M+T carcasses also are pinned. The only piece to furniture that isn't typically pinned but needs the strength is a chair and in that case we all know how chairmakers work very carefully to use dry and wet wood so the final chair dries very tight an solid. Otherwise there will be a problem.

Glue should really be used as a way to keep the parts from sliding apart. THe mechanical strength of the joint - DT or M+T comes from the quality of the fit. If I had for example a .030 gap I would glue in a shim rather than try to rely on gap filling glue.

Re: Hand cut versus machine cut

bob marino

>Joel,

Good points, but couldn't you also use an epoxy glue (with gap filling properties) to compensate for a slight or even not that slight, gap?

Bob

Re: Hand cut versus machine cut

Dan Donaldson

>I would guess that, assuming the same angles, there would not be any difference. The strength of the dovetail is mostly in the direction that puts the wood grain in tension, which is very strong. In that direction, I doubt that the glue has much impact. They would hold just fine with no glue. For the other direction, there is normally so much glue area that, again, it is the strength of the wood that would determine the strength, and I doubt that whether they were cut by hand or machine would have much, if any effect.

This is strictly an opinion without data to back it up ;-)

Joint strength (long)

joel

>no - In a well fitting tenon when you pull it apart there is an audible pop when the suction breaks. THat's how I was taught and once I learned to saw straight it wasn't a big deal to get. That's why I was taught to pare the sides to the mortise to the scribe line also - it allows very precise sizing.

When I screw up I glue a piece of veneer to the tenon and use a shoulder plane to get back to the right size. I use normal glue (formally titebond but now liquid hide glue) and if there were no minor gaps the joint would be totally starved.

THe problem with all glues is that they will loosen to due wood movement over the years and the more thick and rigid the chunk of glue (read epoxy) the worse the problem will become.

In a properly fitted and sized mortise torque on the joint is prevented by the shoulders of the joint and the overall size of the tenon. If the joint is in tension it will eventually fail unless pinned in place. (glues work in shear but not extremely well for a long long time)

Wood movement will of course slowly make the joint deteriorate but not that much. THe real problem with wood movement is if a gap shows up on the side where the rail and stile are joined. This give space for torque to move the joint and break the glue join. The solution is draw boring.

A lot of how a joint is designed is dependant on what it's for. big doors were always drawbored because doors are in motion and far from the hinge there is tons of of torgue on the joints.

On the other extreme the dust partitions in drawers are usually M+T together but not even glued - because a little movement is fine and they don't have any structual role. (rails in a carcass that are supposed to be structural are usually dovetailed in)

Re: various glue's strength in mortise joints

Bill Tindall, E. TN

>Are you sure of your numbers for hand cut? What we would call a good joint is about 0.003 gap (I have measured such with micrometer to satisfy my curiosity) It presses together snug and holds itself together during assembly. A 0.01 gap is so loose that assembly is a problem. I don't think many of us would accept 0.010 fit in a M&T or dovetail joint. It is so sloppy that alignment becomes a problem.

The performance of a glue for gap filling will be a function of its shrinkage during cure and the glue's inherent strength. I suspect that the inherent strength of all the synthetic glues is greater than hide but the shrinkage factor for PVA should be fatal for gap filling. Cyanoacrylate and epoxy have little shrinkage during cure. I don't know about hide-liquid or meltable. When the hide glue pot cools does it dimple in the center, indicating shrinkage?

"pare the sides"

Brent Langdon, Sterling VA

>I am far from and expert on hand chopping mortises as I have only cut a handful with a 1/2" mortising chisel (3/8" and 1/4" ones are on my tool wish list). The way that I was taught (a.k.a. read about on the Internet) was that the width of the chisel determined the width of the mortise. Seems like paring the sides after chopping would screw up the dynamics.

- Brent

Re: "pare the sides"

Dan Donaldson

>You cut the tenon to fit, so the exact mortise width does not really matter.

Re: "pare the sides"

joel

>THe way I was taught (and it's a very old technique) was to set the mortise guage to 1/32 larger than the width of the chisel. THen chop - no need to be need - sloppy is quicker. than with a wide paring chisel put the chisel on the scribe lines, press down, and pare off the 1/64 to the line. You end up with a dead smooth tight mortice.

The other advantage of this technique is that if you make your tenons on a table saw you do the tenons first. Gauge the mortises to the actual size of the tenons , chop using the closest size chisel, then pare to final perfect dimension.

Clarification

Adam Cherubini, NJ

>Hi Brent,

Just to clarify. I think my mortises are beautiful. Its the tenons that are rough. I don't ever intend to plane or pare either mortise or tenon. If a joint won't go together I pare the tenon until it does.

Regarding fit up, Bill may be right. I may not be correct about how loose my joints are. I do know that they are often irregular. I'm just making .030" with my fingers and looking at it. I could easily be off that much in one area and still be beating my joints together. hmmm...I think so.

FWIW, I never have a joint that makes this precision suction thing. When my joints are really tight, they go together and I can't get them apart without fear of breaking something. They never have that precision feel.

Adam

Re: various glue's strength in mortise joints

Christopher Fitch @ Memphis

>This whole thread is pretty good stuff: from the initial data on M/T strength tests to comments on steps for making the joints.

On a side note, it's things like this that I would like to see in some woodworking mags...as opposed to yet another Tool Test.

Another thing I would like to see is discussions on design or why someone might choose a certain joint over another or other such furniture design questions.

*sigh*

Re: various glue's strength in mortise joints

Scott Post

>Adam, I think you're off by a factor of 10 in your measurements. I've never had a joint approaching 0.030" slop. A typical M&T is a slip fit where it isn't pounded together but will not fall apart (without glue) either. A gap of 0.030" is way too much. That's more than a saw kerf wide. A quickly cut joint done with hand tools will still be accurate somewhere in the neighborhood of 0.005" or less.

I know we don't like to use machinist type measurements in woodworking, but now and again I'll pull out my calipers and have a look and I'm always surprise just how accurate hand tools can be.

Depends some on the mortise, too

Bill Houghton, Sebastopol, CA

>Lock mortises are best cut by the drill and pare method - there's so little width of wood left, especially in interior doors, that using a mortise chisel would probably blow out the remaining wood.

A very wide paring chisel (mine is either 1-3/4" or 2", can't remember now) is very helpful here.